Establishment of an Improved Mouse Model of Endometriosis Allowing the Non-Invasive Assessment of Implantation, Progression and Regression of Human Endometriotic Lesions

Objective: To develop an improved mouse model of endometriosis to allow the non-invasive and dynamic monitoring of human endometriotic lesion implantation and regression. Design: Develop a murine non-invasive experimental model for endometriosis by implanting green fluorescent protein (GFP) expressing fragments of human endometrium into immunocompromised mice. Materials/Methods: We developed an improved, non-invasive animal model for endometriosis. To this end, we introduced the GFP gene into 1–2 mm3 fragments of proliferative phase endometrium, by adenoviral infection in vitro. Ten fragments of GFP-expressing endometrial tissue were then injected either subcutaneously or intra-peritoneally into ovariectomized Nude mice, previously implanted with an estradiol-releasing pellet. Living transplanted mice were periodically examined using a light emitting box allowing GFP excitation. Image analysis allowed quantification of fluorescence intensity and size of endometriotic lesions. In a second set of experiments, monitoring of lesion evolution in untreated or progesterone-treated animals was performed by serial in vivo imaging of animals. Number of lesions, their size and fluorescence intensity, as well as time of complete disappearance were quantified and compared between groups. Results: One day after adenoviral infection, endometrial tissue was shown to be highly fluorescent both by microscopic observation and flow cytometric analysis. GFP-expressing endometrial tissue transplanted to Nude mice was able to form endometriotic lesions. Importantly, fluorescent endometriotic lesions were readily observed non-invasively in subcutaneous and intra-peritoneal sites. Lesion evolution could be followed in the same animal over a 3 week period, and their size and fluorescence intensity could be measured. These two parameters spontaneously decreased with time, likely due to GFP dilution among dividing cells. We postulated that decrease in size and fluorescence intensity of lesions would be more pronounced in mice undergoing endometriosis medication. To make the proof of concept that this model could indeed be used for the pre-clinical testing of new drugs targeting endometriosis, we treated Nude mice exhibiting fluorescent lesions with progesterone. Mice treated with progesterone had a faster and more intense decrease in the number and size of lesions, as assessed by non-invasive imaging. Conclusions: The easier identification of human endometriotic lesions in animals is a major strength of this improved model. Moreover, it allows the visualization of lesion evolution after treatment in the same animal, non-invasively, reducing time, cost and number of animals involved when assessing the efficacy of drugs directed against endometriosis. The dynamic and non-invasive characteristics of this improved mouse model should make it an unprecedented tool for compound evaluation in a pre-clinical setting of endometriosis. Furthermore, this animal model could have many other applications in endometriosis research, such as evaluating the effect of tested genes on lesion implantation and regression. Supported By: PROCREA BioSciences.